...
首页> 外文期刊>Physical review.B.Condensed matter and materials physics >Variational classical networks for dynamics in interacting quantum matter
【24h】

Variational classical networks for dynamics in interacting quantum matter

机译:相互作用Quantum物质动态的变形古典网络

获取原文
获取原文并翻译 | 示例

摘要

Dynamics in correlated quantum matter is a hard problem, as its exact solution generally involves a computational effort that grows exponentially with the number of constituents. While remarkable progress has been witnessed in recent years for one-dimensional systems, much less has been achieved for interacting quantum models in higher dimensions, since they incorporate an additional layer of complexity. In this work, we employ a variational method that allows for an efficient and controlled computation of the dynamics of quantum many-body systems in one and higher dimensions. The approach presented here introduces a variational class of wave functions based on complex networks of classical spins akin to artificial neural networks, which can be constructed in a controlled fashion. We provide a detailed prescription for such constructions and illustrate their performance by studying quantum quenches in one-and two-dimensional models. In particular, we investigate the nonequilibrium dynamics of a genuinely interacting two-dimensional lattice gauge theory, the quantum link model, for which we have recently shown-employing the technique discussed thoroughly in this paper-that it features disorder-free localization dynamics [P. Karpov et al., Phys. Rev. Lett. 126, 130401 (2021)]. The present work not only supplies a framework to address purely theoretical questions but also could be used to provide a theoretical description of experiments in quantum simulators, which have recently seen an increased effort targeting two-dimensional geometries. Importantly, our method can be applied to any quantum many-body system with a well-delined classical limit.
机译:相关量子物质中的动态是一个难以解决的问题,因为其精确解决方案通常涉及与成分数量呈指数成分的计算工作。虽然近年来,近年来对一维系统得到了显着进展,但对于在更高的尺寸中相互作用的量子模型,已经实现了更大的减少,因为它们包含了额外的复杂性。在这项工作中,我们采用了一种变分方法,其允许在一个更高的尺寸和更高尺寸中进行高效和控制的量子的动态。本文介绍的方法介绍了基于基于古典旋转网络类似于人工神经网络的复杂网络的变分类,这可以以受控的方式构造。我们为这种结构提供了详细的处方,并通过在单维模型中研究量子淬火来说明它们的性能。特别是,我们研究了真正交互的二维格式仪表理论,Quantum Link模型的不足电布动态,我们最近显示了本文中彻底讨论的技术 - 它具有无序的定位动态[P 。 karpov等人。,phy。 rev. lett。 126,130401(2021)]。目前的工作不仅提供了一个框架来解决纯粹的理论问题,而且还可用于提供Quantum模拟器的实验的理论描述,该实验最近看到了靶向二维几何形状的增加的努力。重要的是,我们的方法可以应用于具有良好呈现的经典极限的量子多体系。

著录项

  • 来源
    《Physical review.B.Condensed matter and materials physics》 |2021年第16期|165103.1-165103.17|共17页
  • 作者单位

    Max Planck Institute for the Physics of Complex Systems Noethnitzer Strasse 38 Dresden 01187 Germany;

    University of California Berkeley California 94720 USA;

    Paul Scherrer Institute Forschungsstrasse 111 5232 Villigen Switzerland Department of Physics National Tsing Hua University Hsinchu 30013 Taiwan;

    Max Planck Institute for the Physics of Complex Systems Noethnitzer Strasse 38 Dresden 01187 Germany National University of Science and Technology 'MISiS' Moscow 119991 Russia;

    Max Planck Institute for the Physics of Complex Systems Noethnitzer Strasse 38 Dresden 01187 Germany;

  • 收录信息
  • 原文格式 PDF
  • 正文语种 eng
  • 中图分类
  • 关键词

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号